I recently wrote this article on Seeking Alpha explaining why I find value in measuring the yield-on-cost and earnings-on-cost of my long-term investments. That is because I choose to evaluate my investment successes and failures by judging the ongoing business performance of my holdings rather than evaluating the stock price changes of my holdings. When I outlay capital for an investment, I want to make estimates about [A] what kind of dividends my capital outlay will be generating in five years, and [B] what kind of earnings my capital will represent in five years.
Of course, for some investments I'm more concerned about the future yield-on-cost than I am about the future earnings yield-on-cost. For instance, when I bought BP throughout the past year, I reached the conclusion that the dividend should grow by at least 8% annually for the next five years. At the time of my purchase, the quarterly dividend was $0.48 per share (it has since been raised to $0.54). I expect that quarterly dividend to be around $0.80 by 2017 or 2018. That projection of income growth played an important role in my decision-making process. Like a farmer, I make predictions about my harvest to come when I plant my seeds.
Personally, I think that stable 4-5% dividend yielders like ConocoPhillips (COP), BP, and Royal Dutch Shell (RDS.B) are some of the most underrated stocks in the investment world. It's very hard to find a combination of: (1) a high starting dividend, (2) likely long-term dividend growth, and (3) decent capital appreciation these days, and these oil supermajors are some of the best places to start your research to meet that need.
Sometimes, I'll make an investment decision based on future earnings yield-on-cost (rather than future dividend yield-on-cost). For example, I initiated a meaningful position in IBM on Friday at $190 per share. Yes, the future income component will be nice. IBM is one of the few large-cap companies out there that stands a reasonable chance of raising its dividend by 10% annually for the next five years. However, the starting yield is under 2%. This is an investment that's all about future earnings per share growth.
Maybe I'm living in bizarro world, but I did not think that IBM's earnings were that bad. Really, we're getting upset over a 1% decline in net income and a 5% decline revenue? I do not get the hysteria on this one. If investors are getting this spooked by a minor 5% revenue decline, what are they going to do when they have to handle something terribly bad actually happening to one of their investments?
The joy of reading IBM's annual report is that most of these perceived shortfalls have been accounted for. IBM has long recognized that overall revenue growth will be modest, and that is why IBM has chosen to become a powerhouse stock buyback company. From 2000 to 2012, IBM spent $123 billion buying back stock. Incidentally, share price declines like the 10% fall on Friday actually help long-term IBM shareholders because the company is continuously buying back shares, and if the price stays around $190, it will be able to buy back more shares and accelerate earnings per share growth that becomes increasingly effective as the share price declines.
This does not mean that the revenue shortfall for IBM is a welcome bit of news, but it should be viewed through the prism of long-term investing (heck, IBM shareholders had to tolerate a 40% revenue decline in 1931-1932 relative to the 1927 period, and July 1932 would have been the best time to buy shares in the company's history). IBM is one of two dozen S&P 500 companies that generates over 10% of its revenue in Japan, and the yen has fallen over 20% relative to the dollar since the end of 2012. IBM has a $105 billion pension obligation, and right now, the discount rate is artificially low because of the prevailing interest rates. Once interest rates rise to 3-5%, the discount rate will rise (because the fixed income portions of the pension will be throwing off more income), and this will take the pressure off IBM.
Long story short, IBM still seems on pace to hit its target of $20 per share in earnings by 2015 (as an aside, IBM's projection of $20 per share excludes non-operating expenses, so the number you will hear in 2015 will likely be closer to $18.50 per share). In 30 or so months, I could reasonably be looking at an earnings yield on cost of 10.5% if you exclude non-operating expenses, or 9.73% if you don't. When IBM is generating $19-$20 per share, I'll expect it to trade at about 13x earnings, or $247 to $260 about 30 months from now (assuming that the market is rationally valued at that time).
When I contemplate an investment, I like to make rough projections about how much wealth will come courtesy of current dividends, future dividend growth, and capital appreciation based on a rational valuation of earnings growth. In the case of BP, I got a nice starter yield of 5% that I expect to grow by around 8% in the next five years, and most of my total return projections hinge on the dividend. In the case of IBM, most of my projections are based on earnings growth (largely fueled by a buyback) that should eventually be recognized by Mr. Market with a 12-14x earnings multiple.
Naturally, this story is not meant to encourage you to buy BP or IBM. I do not know anything about your risk tolerance, time horizon, general temperament, or long-range investing goals. Rather, I wanted to share with you part of my investing process in the event that you might find it useful. When I consider an investment, I try to make realistic five-year projections about the earnings growth and dividend growth relative to the price I pay for the security and the starting dividend yield. My predictions may not always be right, but it is usually helpful to explicitly outline my business performance expectations from the stock beforehand so that I can see if my individual holdings are meeting expectations.